- Use @functools.lru_cache(maxsize=None) decorator for remove_accents()
(Makes everything 30%-40% faster)
- Add sorting option to tabstatistics
- add enchant2 and enchant2-aspell and aspell-en to BuildRequires:
- Convert user database context to lower case and accents removed
while reading training data
- Make the context in the database case insensitive and accent insensitive
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/256)
- Test cases for the sqlite database
- Allow cleanup_database() to cleanup an in memory database when doing unit tests
- Allow only lower case user shortcuts
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/255)
- Add function to return number of rows in the database for debugging and testing
- Allow to read training data from file when database is empty,
allow reading from .gz files
- Add some extra logging for the emoji tests
- Add 'G-_' to 'G- ' as exceptions which must not trigger commits but
insert into the preedit
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/243)
- Resolves: rhbz#2002199
- Update emoji annotations from CLDR
- Translation update from Weblate (ca, es, it updated)
- Update emoji annotations from CLDR
- Add IBus.KEY_KP_Decimal and IBus.KEY_period as a normal/keypad equivalence pair
- get_string_from_lookup_table_cursor_pos() should return the string in NFC
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/234)
- If an exception happens when trying to play a sound, catch it.
- Change default input method for pa from “pa-inscript2” to “pa-inscript2-guru”
- Make appearance tab narrower if possible (helps especially for the French translations)
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/230)
- Spellchecking the preëdit should only spellcheck the part of the preedit which is a word
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/229)
- Starting a compose sequence while a candidate is selected should
append to the selection
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/175)
- Translation update from Weblate (de, nl, pl, tr, uk all updated to 100%)
- Fix typo in translatable message (Thanks to Yuri Chornoivan)
- Add a few more combining chars for dead keys to the list (Thanks to Matthias Clasen)
- Use the dead keys added recently to ibus as soon as they become available
by an ibus update
(see: 3e2609e68c)
- Make Compose preedit less intrusive, show compose key only in
the middle of the sequence or when it is the only key so far
- Use · U+00B7 MIDDLE DOT instead of ⎄ U+2384 COMPOSITION SYMBOL to display
Multi_key in pre-edit.
- Add an option to play a sound on error (for example invalid compose sequence)
- Don’t include the system compose files unconditionally if a user compose file exists
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/170)
- Add option to choose whether to start up in direct input mode
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/169)
- Change default input methods for Indian locales to inscript2
- Resolves: rhbz#1970626 Ensure enum.Flag's obj._value_ is an integer
(Thanks to Miro Hrončok!)
- Use version.parse from packaging instead of LooseVersion from distutils.version
(To fix a deprecation warning, see: https://www.python.org/dev/peps/pep-0632/)
- Update py-compile to the version from automake-1.16
(To fix DeprecationWarning: the imp module is deprecated in favour of
importlib and slated for removal in Python 3.12;)
- In main.py “import factory” only when the --xml option is not used
Resolves: rhbz#1711646
- Limit the “capitalize” case mode used in auto-capitalization to the
first index in a string
Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/167
- Use sensitivity feature of buttons in the setup tool tab where shortcuts are defined
- Don’t let dialog action area of setup tool expand.
- Improve the setup tool to make it possible to enter multi-line expansions of shortcuts
- Resolves https://github.com/mike-fabian/ibus-typing-booster/issues/158
- Elide extremely long candidates in the lookup table in the middle.
- If a candidate contains newlines, replace the newlines with arrows in the lookup table.
- If a commit string contains new-lines, commit it in several chunks
and forward Return between the chunks.
- Update emoji annotations from CLDR
- Put the remote icon back in emoji-picker.appdata.xml, it was a
false positive by appstreamcli
- Translation update from Weblate (pt_PT updated, 100% complete now)
- Skip the inscript2 test cases if these transliterations are not available.
- Run the voikko tests only for libvoikko version >= 4.3
- Adapt voikko test cases to the newest voikko version
libvoikko-4.3-3, python3-libvoikko-4.3-3, voikko-fi-2.4-3
- Log a warning when language is “fi” and “import libvoikko” failed
- Return an empty list [] in spellcheck_suggest_voikko() when libvoikko was not imported
- Update emoji annotations from CLDR, switching “no” and “nb”
See: https://unicode-org.atlassian.net/browse/CLDR-2698
- Remove icon in emoji-picker.appdata.xml (to avoid warnings
from appstream-util and appstreamcli).
- Use “from unittest import mock” instead of just “import mock”.
- Update emoji annotations from CLDR
- Translation update from Weblate for zh_CN (still 100% complete)
- Add vi-telex to the default input methods for vi_VN locale
- Enable Unicode 13.1 Emoji in emoji-picker by default
- Update emoji data to Unicode 13.1
- Translation update from Weblate for zh_CN
- Use IBus.InputHints.{UPPERCASE_CHARS,UPPERCASE_WORDS,LOWERCASE}
- Add utility to manually test input purpose and input
- Improve handling of input purpose and input hints
- Add compatibility enum classes InputPurpose and InputHints
- Add type hints
- Change default for emoji_unicode_min to 0.0
(fixes emoji omitted from browsing in emoji-picker)
- Add option to automatically capitalize after punctuation
- Don't record in user database when stripped_input_phrase or
stripped_commit_phrase are emty
- When interactively deleting a candidate from the user database,
remove all case modes
- Add 4 more characters ÞĦŊŦ to get special treatment in remove_accents()
- Translation updates from Weblate for ca, es, fr, it, ja, nl, sv
- Fix typo in variable name in do_reset()
- KP_Delete should be handled the same way as Delete
- Translation update form Weblate for de, he, pl, tr, uk
- Prevent also Delete from reopening a preedit when the option
“Arrow keys can reopen a preedit” is off
- When the input is empty, Escape should be passed through,
not inserted into the preedit
- Fix itb_util.tokenize(): if the input is only whitespace,
the return should be an empty list
- Don’t clear context after typing Return, KP_Enter, ISO_Enter
- Improve behaviour of case modes
- Update emoji annotations from CLDR
- Add commands “next_case_mode” and “previous_case_mode”
with configurable key bindings
- Better hot key handling for modifier keys like Shift_L, …
- Use labels 1, 2, 3, … for the lookup table instead of 1., 2., 3., …
- Update emoji annotations from CLDR
- Make translations of 'Edit key bindings for command “%s”' work
- Don't hide emoji-picker.desktop from AppStream (by Gunnar Hjalmarsson)
- Translation update from Weblate for pt_BR, sv (100%)
- Treat characters 'ÅåÄäÖö' as special when matching in the Swedish dictionary
(Resolves: http ://github.com/mike-fabian/ibus-typing-booster/issues/126)
- Update emoji annotations from CLDR
- AppStream tweaks by Gunnar Hjalmarsson
- Translation update from Weblate for ca, de, es, sv
- Add emoji-picker.appdata.xml
- Fix warnings and errors in typing-booster.appdata.xml
- Update emoji annotations from CLDR
- Support %S expansion in include statements in compose files
- Translation update from Weblate for fr, he, hu, zh_CN
- Translation updates from Weblate for de, ja, pl, tr, uk, zh_CN
- Update ibus-typing-booster.pot (Some new translatable strings, tooltips for 2 new buttons)
- Improve matching of keybindings, consider IBus.ModifierType.MODIFIER_MASK
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/94)
- Show the list of dictionaries in the setup tool only if it is more than just ['None']
- If a real dictionary is added in the setup tool, remove dummy dictionary 'None'.
- Add black flag for the special dummy dictionary 'None'
- Add doctests to check that the special dictionary 'None' is handled correctly
- When the list of imes or dictionaries is emptied by the user, set it to 'NoIME' or 'None'
(Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/98)
- Add "Set to default" buttons for the list of input methods and dictionaries
- Remove the special candidate which is shown when a dictionary is missing
- Prevent also BackSpace from reopening a preedit when the option
“Arrow keys can reopen a preedit” is off
- Resolves: https://github.com/mike-fabian/ibus-typing-booster/issues/87
- Resolves: rhbz#1637647
- Translation updates from Weblate for es, tr, he
- Update emoji annotations from CLDR
- Translation updates from Weblate for ca, de, es, fr, nl, pl, uk
- Update emoji annotations from CLDR
- Update emoji data to Unicode 13.0 final
- New option to choose whether spellchecking is done on the preedit
and which colour to use
- Spellcheck typed string in preedit and colour it if it is likely to be misspelled
- More test cases for spellchecking and spellchecking suggestions
- Restructure code for spellchecking and spellchecking suggestions
- Translation updates from Weblate for es, de, pl, he, fr, uk, tr, nl
- Generate README.html and README from README.md
- A thorough upgrade to README.md
Resolves: https://github.com/mike-fabian/ibus-typing-booster/pull/74
- Set button label of google application credentials button correctly
Resolves: rhbz#1793460
- Make si-wijesekera the default input method for si_LK.UTF-8 locale
- Add another test case for libvoikko spellchecking together with
en_GB spellchecking
- Use included fi_FI.dic for Finnish word suggestions
- Add Finnish ispell dictionary fi_FI.dic
- Minor translation updates from Weblate for pl, tr, zh_CN
- Correct release dates of Unicode 12.0.0 and 12.1.0 in emoji_picker.py
- Add workaround if key codes cannot be found for key values
(workaround for big endian platforms)
- Fix race condition in itb_util.xdg_save_data_path()
Resolves: rhbz#1770072
Resolves: rhbz#1713963
Resolves: rhbz#1764520
Resolves: rhbz#1768016
- Another fix for right-to-left languages: fix wrong order of globe emoji
- Small fix in parsing emoji-test.txt which slightly changes the emoji sorting order
- Adapt parsing of the emoji data files to the changes in Unicode 13.0
- Fix emoji test cases for new emoji data files for Unicode 13.0
- Update emoji data to current Unicode 13.0 draft
- Add fa, it, tr to LINGUAS
- Updated Odia translations from Weblate (53.4% translated)
- Updated Czech translations from Weblate (100.0% translated)
- Updated Catalan translations from Weblate (17.1% translated)
- Added Turkish translations from Weblate (100.0% translated)
- Added Italian translations from Weblate (5.2% translated)
- Portuguese (Portugal) translations moved from pt_PT to pt (23.8% translated)
- Update French translations from Weblate (100.0% translated)
- Added Persian translations from Weblate (0% translated)
- Fix display for right-to-left languages in the dictionaries and
input methods tab in the setup tool
- Added Swahili translations from Weblate (30.1% translated)
- Added Traditional Chinese translations from Weblate (0.5% translated)
- Added Portuguese (Portugal) translations from Weblate (23.8% translated)
- Update Hebrew translations from Weblate (100% translated)
- Update Arabic translations from Weblate (100% translated)
- Update Odia translations from Weblate (44.0% translated)
- Update Spanisch translations from Weblate (93.3% translated)
- Update Chinese (Simplified) translations from Weblate (13.5% translated)
- Update Spanish translations from Weblate (92.2% translated)
- Update Catalan translations from Weblate (12.4% translated)
- Add Hebrew translation from Weblate (67.4% translated)
- Add Arabic translation from Weblate (28.0% translated)
- Update Ukrainian translations from Weblate (100.0% translated)
- Update German translations from Weblate (100.0% translated)
- Update Japanese translations from Weblate (35.2% translated)
- Update French translations from Weblate (93.8% translated)
- Add Hungarian translation from Weblate (58.0% translated)
- Add Odia translations from Weblate (2.6% translated)
- Improve compose file parsing to get results containing \" correct
- More test cases for compose sequences
- Change README.md to refer to Weblate for translations instead of Zanata
- Fix display of code points in emoji-picker when running in or_IN.UTF-8 locale
- Quote the tables for the key and mouse bindings in the README.md as code
- Do not hardcode the list of useful m17n input methods
- Use LOGGER.exeption() in exception handlers
- Use python logging module with log file rotation instead of writing to stdout/stderr
- Fix set_preedit_underline() with parameter update_gsettings=True
- If the first candidate is exactly the same as the typed string
prefer longer candidates (Extends inline completions automatically)
- Move README to README.md and use some markdown to make it look better on github
- When showing similar emoji in the lookup table, show the list of keywords
which matched only when debugging is on
- Update emoji annotations from CLDR
- More unittests, restructure test files, move some doctests to unittests
- Add a utility class KeyvalsToKeycodes to find ibus key codes
instead of hardcoding them
- Add ceb to CLDR_ANNOTATION_FILES
